In most patients with aortic regurgitation (AR), aortic valve replacement (AVR) results in favorable left ventricular (LV) remodeling and normalization of the LV ejection fraction (EF). However, some patients with severe AR will not have favorable remodeling and their LVEF will not normalize. The goal of the present study was to determine whether remodeling and clinical outcomes after AVR could be predicted from simple preoperative echocardiographic analysis. A total of 56 consecutive patients with chronic severe pure AR who underwent AVR had preoperative (5 ± 2 days), early postoperative (5 ± 2 days), and late postoperative (328 ± 88 days) echocardiographic data retrospectively analyzed. The LV diameter, The LVEF and stroke volume (SV) were measured. The reduction in LV end-diastolic dimension decreased by 14% (from 65 ± 6 mm to 56 ± 8 mm, p <0.001) early after AVR, with an additional reduction of only 6% late after AVR. More than 2/3 of the overall reduction in end-diastolic dimension was observed the week after AVR. Forty-six patients (82%) had positive early LV remodeling, defined as a 10% reduction in the LV end-diastolic diameter 1 week after AVR. All patients with early LV remodeling had a preoperative SV of ≥97 ml, which was the best predictor of late postoperative LVEF of ≥45% (sensitivity 98% and specificity 100%). Patients with a preoperative SV of ≥97 ml had a markedly greater event-free survival rate (92% vs 13%, p <0.001) at 3 years. In conclusion, in patients undergoing AVR for chronic severe pure AR, preoperative SV is the best predictor of LV remodeling and outcomes.
Most symptomatic patients undergoing aortic valve replacement (AVR) for aortic regurgitation (AR) have symptom improvement after AVR. Others, however, develop progressive congestive heart failure, the most common cause of late postoperative death. In such patients, irreversible myocardial changes presumably occur before or concomitantly with the development of symptoms that lead to the operation. Therefore, the reliable prediction of postoperative left ventricular (LV) systolic function is of critical concern in establishing the appropriate timing of AVR. To detect subclinical LV myocardial dysfunction, the parameters of LV systolic function such as LV ejection fraction (EF) have been widely used. However, it is not uncommon that the preoperative LVEF does not correlate with the postoperative LVEF. Therefore, there is a clinical need to identify a new parameter to predict postoperative LV remodeling, systolic function, and outcomes in patients undergoing AVR for severe AR. The objective of the present study was to determine whether the preoperative stroke volume (SV) could be used to accurately predict early and late LV remodeling and outcomes in patients with chronic severe pure AR.
Methods
We studied patients with chronic severe pure AR who had undergone AVR from January 2005 to June 2010 in our institution. Patients with acute AR, coronary artery disease (coronary artery stenosis >50%), atrial fibrillation, aortic stenosis, significant mitral disease (more than mild disease), previous aortic valve surgery, and previous or associated mitral valve replacement or repair were excluded. A total of 56 patients were included in the present study. The study was approved by the local hospital ethic and scientific committees. The patients were separated into 2 groups according to the presence or absence of early LV diastolic remodeling. Early diastolic remodeling was defined as an early postoperative LV diastolic dimension reduction of ≥10%. Postoperative cardiac events were defined as rehospitalization for congestive heart failure or cardiovascular-related death. The median duration of follow-up was 2.6 ± 2.0 years. Two-dimensional and Doppler transthoracic echocardiographic examinations with commercially available echocardiographic systems (Sonos 5500 or 7500, Philips Medical Systems, Amsterdam, The Netherlands) were performed 5 ± 2 days before AVR and 5 ± 2 days (early) and 328 ± 88 days (late) postoperatively. The degree of AR (grade 3+ to 4) was determined by analysis of color flow Doppler. The heart diameter was measured using 2-dimensional echocardiographically guided M-mode, as previously described. The LV end-diastolic and end-systolic volumes and LVEF were determined using the modified biplane Simpson method. End-systolic wall stress was also calculated, as previously described. The SV was calculated using the outflow tract diameter measured at the base of the aortic leaflet in the parasternal long-axis view and the pulsed Doppler velocity integral obtained at the same level in the 5-chamber view.
Continuous variables are presented as the mean ± SD; discrete variables are presented as the frequency distribution. The mean values were compared using the 2-sample independent and paired t test, and categorical variables were compared using conventional chi-square testing. The correlations are reported with Pearson’s coefficient. All time-to-event distributions were estimated with the Kaplan-Meier methods. All reported time-to-event comparisons were made using the log-rank test.
Results
The etiology of AR was a bicuspid valve in 23 patients (41%), rheumatic/degenerative disease in 16 patients (29%), aortic dilation in 15 (27%), and miscellaneous causes in 2 patients (3%). Thirty-two patients (57%) required concomitant ascending aortic surgery. Of the 56 patients, 46 (82%) had early significant LV remodeling (10% LV diastolic diameter reduction ≤2 weeks after AVR) and 10 (18%) had no early significant LV remodeling. No significant difference was found between patients with and without early remodeling in terms of preoperative demographic and clinical data ( Table 1 ). In contrast, the echocardiographic characteristics were all different between the patients with and without early LV remodeling.
| Variable | Early Remodeling (n = 46) | No Remodeling (n = 10) | p Value |
|---|---|---|---|
| Age (years) | 53 ± 2 | 61 ± 4 | 0.06 |
| Men | 35 (76%) | 8 (80%) | 0.9 |
| Mean systolic pressure (mm Hg) | 125 ± 3 | 127 ± 8 | 0.8 |
| Mean diastolic pressure (mm Hg) | 66 ± 1 | 68 ± 3 | 0.5 |
| New York Heart Association functional class | 3.0 ± 0.3 | 3.0 ± 0.4 | 0.8 |
| Body surface area | 2.0 ± 0.2 | 2.0 ± 0.3 | 0.9 |
| Left ventricular ejection fraction (%) | 47 ± 10 | 32 ± 6 | <0.001 |
| Left ventricular end-diastolic diameter (mm) | 65 ± 6 | 69 ± 8 | <0.001 |
| Left ventricular end-systolic diameter (mm) | 45 ± 7 | 56 ± 10 | <0.001 |
| Stroke volume (ml) | 130 ± 27 | 70 ± 12 | <0.001 |
| Left ventricular end-diastolic volume (ml) | 134 ± 31 | 211 ± 59 | 0.002 |
| Left ventricular end-systolic volume (ml) | 82 ± 28 | 162 ± 53 | <0.001 |
| Left ventricular diameter/wall thickness | 6 ± 1 | 7 ± 1 | 0.001 |
| Left ventricular systolic/wall thickness | 4 ± 1 | 5 ± 1 | <0.001 |
| Systolic wall stress | 128 ± 32 | 185 ± 42 | <0.001 |
On simple linear regression analysis, the preoperative SV (r = 0.44, p <0.001), LVEF (r = 0.34, p <0.01), and LV end-diastolic dimension (r = 0.26, p = 0.06) correlated with early LV diastolic remodeling. The preoperative presence of a SV of ≥97 ml had the best performance for the prediction of early LV remodeling compared to the preoperative LVEF and indexed LV diastolic or systolic diameter ( Figure 1 ). The reduction in the LV end-diastolic dimension decreased by 14% (from 65 ± 6 mm to 56 ± 8 mm, p <0.001) early after AVR, with an additional reduction of only 6% (relative to preoperative values) to 52 ± 8 mm late after AVR. Thus, 69% of the overall reduction in end-diastolic dimension observed during the long-term postoperative evaluation occurred within 2 weeks of AVR. A reduction in the LV end-systolic diameter decreased by 9% (from 47 ± 8 mm to 43 ± 9 mm early after AVR, p <0.001) with an additional reduction of 9% to 39 ± 10 mm late after AVR. By definition, patients with significant early remodeling (≥10% LV end-diastolic diameter early after AVR) had more important LV end-diastolic diameter reduction than patients without significant early remodeling ( Figure 2 ). LV end-systolic reduction was also more important in the group of patients with significant early remodeling ( Figure 2 ). In all patients, the LVEF decreased by 22% from 46 ± 11% to 36 ± 11% early after AVR (p <0.001) with an additional increase of 11% (relative to the preoperative values) to 51 ± 13% late after AVR. The patients with significant early remodeling also had a decrease in LVEF early after AVR; however, the LVEF increased significantly late after AVR ( Figure 3 ). In patients without significant early remodeling, the LVEF decreased significantly late after AVR compared to the preoperative values (32% vs 26%, p = 0.03). The preoperative presence of a SV of ≥97 ml had the best performance for the prediction of a LVEF of ≥45% late after AVR ( Table 2 ). The patients with a preoperative SV <97 ml had a markedly lower cardiac event-free survival at 3 years compared to those with a preoperative SV of ≥97 ml ( Figure 4 ). The outcomes for patients with a SV of <97 ml are presented in Table 3 .
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
